BREVIA

and Flock House (FHV; Nodaviridae). The latter is unrelated to DCV and CrPV and is patho- genic in adult flies (7), although natural infections and Virus have not been reported. Like DCV, both CrPVand FHV induce rapid mortality when injected into Protection in Insects adult Drosophila. All Oregon RC flies infected with Wolbachia and CrPV died within 17 days postinfection (Fig. 1C). In contrast, the Wolbachia- Lauren M. Hedges, Jeremy C. Brownlie, Scott L. O’Neill, Karyn N. Johnson* free Oregon RC flies died within 7 days of in- olbachia pipientis are maternally trans- We compared the survival of flies infected fection. Similarly, Wolbachia-free flies challenged mitted, Gram-negative, obligate in- with DCV in the presence or the absence of with FHV died within 8 days of infection, whereas W tracellular bacteria found in filarial Wolbachia infection (Fig. 1 and fig. S1) (5). In 26 days postinfection only 35% of the Wolbachia- nematodes, crustaceans, arachnids, and at least flies from the standard laboratory strain, Oregon infected flies had succumbed to FHV-induced 20% of all insect species. Many Wolbachia bac- RC, Wolbachia infection delayed DCV-induced mortality (Fig. 1D). These results indicate that the teria increase their prevalence in populations by mortality compared with Oregon RC flies cured of antiviral effect observed in Wolbachia-infected manipulating host reproductive systems (1). In- Wolbachia infection (Fig. 1A). The delay in mor- Drosophila functions to protect flies from diverse sects are also commonly infected with , tality corresponded with a delay in virus accumu- RNA viruses. and, considering the shared intracellular location, lation in Wolbachia-infected flies (fig. S2). The Typically Wolbachia manipulate host repro- it is possible that Wolbachia may influence the experiment was repeated with the fly strain w1118 ductive systems to increase the number of infected outcome of virus infection in an insect host. with similar results observed (Fig. 1B). The sur- hosts within a population. However, Wolbachia is commonly in- vival curves of Oregon RC and w1118 Wolbachia- strains that infect D. melanogaster do not induce fected with Wolbachia and is a powerful model free flies were similar to those of two wild-type these parasitic traits under field conditions at levels for studying host-pathogen interactions and anti- laboratory populations (Champetières and Oregon sufficient to invade host populations (8). Theory viral responses (2). Drosophila C virus (DCV), a R) that are naturally uninfected with Wolbachia predicts that in the absence of strong reproductive member of the family, is a natural (compare Fig. 1, A and B, with fig. S1). Oregon RC parasitism Wolbachia should confer a fitness benefit 1118

pathogen of D. melanogaster and is found in 30 to and w flies are infected with two closely related to the host, but for D. melanogaster no such benefit on October 30, 2008 40% of both laboratory and wild-caught popula- strains of Wolbachia, wMelCS and wMelPop, has been identified in nature (8). Because both tions (3, 4). Infection of adult Drosophila with respectively (6). These results indicate that these DCV and Wolbachia are common in wild Dro- DCV by injection can result in 100% mortality strains of Wolbachia, in different genetic back- sophila populations, the association of Wolbachia within 3 to 4 days. Although variation in sus- grounds of Drosophila, have an antiviral effect. with a robust antiviral effect may confer a positive ceptibility of fly strains to DCV-induced mortality Two further viruses were tested with use of the selective advantage to flies. If generalized, the an- has been recorded (3), the underlying basis for this survival bioassay: paralysis virus (CrPV; tiviral protection associated with Wolbachia infec- variation has not been determined. Dicistroviridae), a natural Drosophila pathogen, tion might be exploited in future strategies to reduce insect-transmitted diseases. A B 100 www.sciencemag.org 100 References and Notes 1118 80 ORC-w w w 1. S. L. O'Neill, A. A. Hoffmann, J. H. Werren, Influential 80 1118 Passengers: Inherited Microorganisms and Arthropod ORCT w T 60 Reproduction (Oxford Univ. Press, Oxford, 1997). 60 2. S. Cherry, N. Silverman, Nat. Immunol. 7, 911 (2006). 40 3. G. Brun, N. Plus, in The Genetics and Biology of 40 Drosophila, M. Ashburner, T. F. R. Wright, Eds. (Academic – 20 20 Press, New York, 1980), vol. 2D, pp. 625 702. 4. K. N. Johnson, P. D. Christian, J. Gen. Virol. 79, 191 (1998). Downloaded from 0 0 5. Materials and methods are available as supporting material on Science Online. 02468101214 0 2 4 6 8 10 12 14 6. M. Riegler, M. Sidhu, W. J. Miller, S. L. O'Neill, Curr. Biol. C D 15, 1428 (2005). 100 100 7. X. H. Wang et al., Science 312, 452 (2006); published Survival (%) 80 ORC-w online 22 March 2006 (10.1126/science.1125694). 80 8. A. A. Hoffmann, M. Hercus, H. Dagher, Genetics 148, ORCT 60 60 221 (1998). 9. We thank M. Riegler for advice and E. McGraw and ORC-w 40 40 F. Pringle for critical reading of the manuscript. This ORCT work was supported by funding from The University of 20 20 Queensland, Australian Research Council, and the Foundation for the National Institutes of Health through 0 0 the Grand Challenges in Global Health Initiative. 0 2 4 6 8 1012141618 0 2 4 6 8 10 12 14 16 18 20 22 24 26 Supporting Online Material Days postinfection www.sciencemag.org/cgi/content/full/322/5902/702/DC1 Materials and Methods Fig. 1. Infection with Wolbachia protects flies from virus-induced mortality. The data shown represent the Figs. S1 and S2 mean of triplicates, and the bars indicate standard error. The survival curves were significantly different 26 June 2008; accepted 29 August 2008 for Wolbachia infected versus uninfected flies (Kaplan-Meier analysis, P <0.0001ineachcase).(A) 10.1126/science.1162418 Comparison of the survival of Wolbachia-infected (ORC-w) and uninfected Oregon RC (ORCT) flies after 1118 challenge with DCV. (B) Comparison of the survival of Wolbachia-infected(w)anduninfected(T)w School of Integrative Biology, The University of Queensland, flies after challenge with DCV. (C)ComparisonofthesurvivalofWolbachia-infected (ORC-w) and Brisbane 4072, Australia. uninfected Oregon RC (ORCT) flies after challenge with CrPV. (D)Comparisonofthesurvivalof *To whom correspondence should be addressed. E-mail: Wolbachia-infected (ORC-w) and uninfected Oregon RC (ORCT) flies after challenge with FHV. [email protected]

702 31 OCTOBER 2008 VOL 322 SCIENCE www.sciencemag.org